Regarding Are of Interest
shubhayanroychowdhury opened this issue · 8 comments
Hi @Alex-Lewandowski @khogenso @piyushrpt
Can you explain me the new modified part of area of interest section! It is quite difficult to find out the area of interest.
We now display the latitude and longitude along the axes, instead of the x, y indices of the raster. We've also added gridding to help identify specific locations.
The "Max area covered by stack" contains pixels covered by at least a single scene in the stack. "Common area covered by data stack" contain pixels that are covered by every scene in the stack.
We switched to displaying a basemap instead displaying the merged output of many amplitude products from a stack. This was done to support very large time series and SBAS stacks.
Is it the basemap that you have difficulty with? We are somewhat limited in the selection of supported, free basemaps but can investigate alternative options.
All our code is open source and we happily accept Pull Requests if you have an idea for an improvement. The AOI selector is part of opensarlab-lib
Hi @Alex-Lewandowski
Yes, actually I have some difficulties with basemap. Is this possible to show the map like this!
Also i want to address one issue. When I was applying atmospheric correction step. After atmospheric correction, when I download the velocityERS5.tif file and open it in ArcGIS or SNAP. I got a dem like map instead of actual velocity map.
Thank you for your time and consideration
Hi @shubhayanroychowdhury, Thank you for the additional details. We will explore alternative basemap options and look into why you are ending up with a DEM in place of the expected velocity map. I am adding this to our backlog, and we will try to tackle it within the next several weeks.
Hi @Alex-Lewandowski
Thank you so much for your consideration. Actually I am working on paper where I have used OpenSARLab for time series analysis purpose. So it will be very helpful for me. Please look into it.
Thank you again.
Hi @Alex-Lewandowski
One thing I would like to mention is that, in the past, I was using the Insar_analysis environment for time series analysis. I got this error.
Now, I am using the osl_mintpy environment. The errors are solved. However, when I use the osl_mintpy environment, I do not get any results, whereas previously, when I used the insar_analysis environment, I obtained results. Can you tell me why I am not getting results using the osl_mintpy environment?
Hi @shubhayanroychowdhury
It's a little difficult to determine the cause without some more information on the data set you are working with and a look at the notebook output from a failed run, but a possible source of the trouble does come to mind, given that you have switched between a few updated environments.
In MintPy=1.5.2 a change was introduced that automatically projected HyP3 data to lat/lon. Previously, we had been projecting the data in the data prep notebook. Now, doing that caused an error when MintPy tried to as well. So, we stopped projecting to lat/lon in the prep notebook to let MintPy handle it. Unfortunately, there was an additional issue in that MintPy saved the reference point in the original projection, not lat/lon, which broke it's ability to correctly identify raster pixels in x/y. So, we reverted our previous fix, once again projecting in the prep notebook, and pinned the environment to MintPy=1.5.1.
I suspect, you got caught up in that mess, and your data are now possibly in the wrong projection with an improper reference point saved in your MintPy project. I recommend starting the entire workflow over, starting with redownloading and preparing the data.
This is just my best guess at a solution without knowing more about the data and the specific errors you encountered. If this solution doesn't work, please send a copy of your run notebooks, with the saved output in place.
Additionally, I just noticed that MintPy=1.5.3 released last week. Perhaps it contains a fix and we can stop pinning an older version. We will look into that.
Hi @Alex-Lewandowski
Thank you so much for your guidance. Yes, MintPy latest version has been released last week and now i get a results.
Hi @Alex-Lewandowski
Please look into the issue with atmospheric correction. Today I have processed the data with atmospheric correction. After atmospheric correction, when I download the velocityERS5.tif file and open it in ArcGIS. I got a dem like map again instead of actual velocity map.
Thank you for your time and consideration.